The immediate goal of these studies is to examine a novel mechanism of B- lymphocyte tolerance that involves the negative selection of antigen receptors, rather than the cells expressing them. Preliminary data is presented suggesting that encounter with autoantigen can induce the rearrangement of light chain genes in immature sIgM+ cells of the bone marrow. This was observed in transgenic mice bearing rearranged, functional Ig genes encoding an antibody specific for MHC class I alloantigens, in which the effect of different forms of autoantigen on B- cell development could be assessed. This proposed mechanism for B-cell tolerance in immature B-cells is termed the Receptor Editing hypothesis. The Specific Aim of this proposal is to test the Receptor Editing hypothesis. This hypothesis proposes that in autoreactive, immature B lymphocytes sIgM crosslinking by self antigens induces secondary immunoglobulin gene rearrangements that can alter their receptors and render them non-autoreactive. This problem will be approached by measuring the recombinase gene expression, Ig gene DNA rearrangements, and lifespans of immature, sIgM+ B-cells in the presence and absence of membrane bound autoantigen in vivo and in vitro.GRANT=R01AI17416 Rocky Mountain spotted fever (RMSF), caused by Rickettsia rickettsii, is considered the most severe of the human rickettsioses. Usually transmitted by the bite of a tick, the organisms are introduced directly beneath the skin during a blood meal where they invade and destroy the endothelial cells of small blood vessels. Our laboratory has established an in vitro model to study different parameters of rickettsiae-host interaction including the mechanisms of injury to endothelial cells infected by this organism. An understanding of the mechanism(s) of cell injury caused by R. rickettsii could significantly enhance existing knowledge of the pathogenesis of RMSF, and could provide for more specific therapeutic management of severe forms of the disease. The proposed study will be carried out in cultured human endothelial cells infected by the virulent Sheila Smith strain and the avirulent Iowa strain of R. rickettsii, and for comparison purposes, the virulent Breinl and the avirulent Madrid E strains of R. prowazekii. Current evidence from our laboratory strongly suggests that R. rickettsii-induced endothelial cell injury is mediated by oxygen radicals. Previously, it was assumed that the injury was initiated during intracellular growth of the organism. Now, however, since we have detected superoxide radical in large amounts in the culture supernatants during internalization of the rickettsiae into endothelial cells, we postulate that injury is initiated as a result of an aborted phagocytosis of the organism, that results in an oxidative burst. Rather than killing the rickettsiae, this oxidative burst, uncharacteristically, appears likely to lead to the death of the endothelial cell. Superoxide released at the surface of these cells during internalization is converted to toxic H202 by extracellular superoxide dismutase associated with the glycocalyses of endothelial cells. H202 readily passes through the plasma membrane and is capable of directly causing peroxidation of membrane lipids, or it may react with additional superoxide to produce the even more toxic hydroxyl radical. Catalase and pyruvate are known scavengers of H202 and protect endothelial cells from hydrogen peroxide generated from superoxide. The hypothesis that R. rickettsii causes abortive phagocytosis leading to early endothelial cell death will be tested by examining cells for reduced intracellular and extracellular peroxide levels following pretreatment with these scavengers, measuring intracellular glutathione and glutathione peroxidase levels, examining the cells by EM for dilatation of the E.R., and comparing them as above to the avirulent Iowa strain and the virulent Breinl and avirulent Madrid E strains of R. prowazekii, which appear not to cause oxidative injury.